yay.
> The changes are so small and so obviously
> have to work that it would be ridiculous not to have this - the only
> question is whether I'll try to make it a "copyfd()" system call instead,
> falling back on read+write when I can't use the page cache directly. I
> suspect I won't.
Please reconsider this? There are some things that do a LOT of
network-to-network copies (i.e. proxy servers), that would benefit
from skipping the extra memory copy. And I dare say you could get some
really neat speedups with re-using the skbuff structure in some
fashion.
Ideally just rename the call 'copyfd' and return -ENOTIMPLEMENTED for
the case where the source FD isn't a file/disk.
Either way, it would be a little sad to rule out the possibility of
other XtoY optimizations at this early stage.
Michael, running some very very busy large squids and watching his
idle time sit at 2%... :)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu